Build Your Own Software or Buy an Off-the-Shelf Tool? The Real Math

You're torn between building your own software and picking up a ready-made tool off the market. It's one of the most structuring decisions for your business, and yet it often gets made on a whim, one spreadsheet too many, or a subscription invoice that stings. I'm going to give you the real math, the one I use with the Swiss SMEs I work with. And I'll be honest on both sides, because it's in my interest that you make a good decision, not that you order a poorly scoped project from me.

Let's start with the truth that stings a little in my line of work.

Most of the time, the off-the-shelf tool is enough

I'll say it plainly because it should be reassuring: in most cases, an existing tool is the right choice.

Bexio for your invoicing and accounting, Notion or ClickUp for internal organization, a CRM like Pipedrive for your sales tracking, a Shopify to sell online. These tools cost a few dozen francs a month, they're maintained by entire teams, updated continuously, and they cover 80% of what an SME needs.

Why reinvent that? You gain nothing from owning your own invoicing software. Nobody will buy your services because your accounting runs on custom code. It's time and money burned for a result Bexio already does better.

Rule number one: if an off-the-shelf tool does the job at 80% and the missing 20% costs you nothing serious, take the tool. Adapt your process to the tool, not the other way around. It's less glamorous, but it's more profitable.

Custom isn't always the right answer. A poorly scoped project costs everyone dearly, you most of all.

When building your own software becomes justified

Now, the other side. There are situations where building your own tool isn't a whim but a genuine business decision. I see them regularly.

1. When you spend your time working around the tool

You have an off-the-shelf SaaS, but your team has rigged it up with three Excel spreadsheets on the side, copy-pasting between two apps, a shared file someone updates by hand every Friday. The signal is the workaround. If your real process lives half outside the tool, it means the tool doesn't fit the way you work.

That wasted time has a cost. Two hours a week per employee, at CHF 60 per loaded hour, across five people, quickly adds up to more than CHF 30,000 a year going up in smoke. At that level, software built for your process pays for itself.

2. When the software IS your competitive edge

This is the strongest case. If the way you manage a flow, calculate a price, coordinate your teams, or serve your clients is what sets you apart from the competition, then that process shouldn't be locked inside a generic tool your competitors use too.

Here, custom becomes your asset. It belongs to you, it encodes your know-how, and it builds a barrier no one can copy by subscribing to the same SaaS as you.

3. When per-user subscriptions explode

A tool at CHF 40 per user per month is painless with five people. At forty people, it's CHF 19,200 a year. At eighty people, nearly CHF 40,000, every year, without ever owning anything. Custom software has a higher upfront cost, but once it's paid for, it belongs to you, and its bill doesn't mechanically climb with your growth.

The decision framework, criterion by criterion

Here are the six criteria I review with every SME. Grab a sheet of paper and answer honestly for each one.

Total cost, not upfront cost

The classic mistake is to compare the sticker price of a subscription with the quote for custom development. That makes no sense. Compare the total cost over three to five years.

On one side, the subscription that swells with the number of users and the tier upgrades. On the other, a heavier upfront cost, plus a maintenance budget, but an asset that belongs to you. If you want to put realistic figures on the custom side, I broke it down in my article on the price of custom SaaS. Note too that VAT at 8.1% applies on both sides, so it doesn't change the trade-off.

Fit with your process

What percentage of your way of working does the off-the-shelf tool cover? At 90%, take the tool without hesitation. At 50%, you're going to suffer. Be honest: is your process really special, or is it just a habit you could adapt?

Dependence and ownership

With an off-the-shelf SaaS, you depend on the vendor. They can raise their prices, change their terms, get acquired, shut down a feature you use every day, or disappear. Your custom software, you control it. That's no small thing when the tool sits at the heart of your business.

Data and compliance

Where does your data live? For many SMEs in French-speaking Switzerland, the question of the revised nLPD and Swiss hosting is real, especially in healthcare, finance, or legal. Some foreign SaaS products store everything outside Switzerland. With custom software, you choose where your data is hosted and you control who accesses it.

Timeline

Here, the off-the-shelf tool almost always wins. You subscribe, you configure, you're running in a week. Custom development, even well run, is measured in months. If you need a solution tomorrow morning, take the tool, even if you replace it later.

Scalability

Is your need stable or in motion? A fixed tool suits you if your process doesn't change. But if you know you're going to add flows, connect other systems, evolve your business logic every quarter, software whose code you control follows your growth instead of holding it back.

The simple table I use

To decide quickly, I sum it all up in one question per column.

Take the off-the-shelf tool if: it covers more than 80% of your need, your process isn't your competitive edge, you want to start right away, and the subscription cost stays reasonable given your team size.

Build your own software if: you're constantly working around existing tools, your process makes your difference, your subscriptions become a serious cost line, you have strong data requirements, and your need is going to keep evolving.

And there's a third path I rarely forget to mention: start with an off-the-shelf tool, and build custom only for the part that truly makes your difference. You keep Bexio for accounting, and you build only the business module that belongs to you. It's often the best of both worlds, and the wisest financially.

How I frame this with my partners

When an SME reaches out to me for a project, I don't assume we need to build. My first reflex is to check that an existing tool wouldn't do the job. If it would, I tell you, even if it means I don't take the project.

Custom becomes worthwhile once we've established together that your process has genuine value of its own, that off-the-shelf tools force costly contortions on you, or that owning your data and your business logic is strategic. At that point, a technical partner who builds with you, over the long term, makes sense.

If you want to understand concretely what custom development looks like, from idea to production, I wrote a complete guide to building a custom SaaS platform. It'll give you a clear view of the steps, the technical choices, and the pitfalls to avoid.

In short

The right instinct is neither "I build everything" nor "I subscribe to everything." It's to lay out the real math: total cost over several years, fit with your process, dependence, data, timeline, scalability. Often the off-the-shelf tool wins, and that's perfectly fine. Custom is justified when the tool becomes your competitive edge, or when you spend your time working around it.

If you think you're in that second case, or if you just want an honest opinion on your situation, let's talk it through. I help you decide before writing any code, and build only what's worth building. Discover my work in custom SaaS in Switzerland and write me a few lines about your situation.